Voice-recognition software enables a computer to translate spoken words into digital format–that is, into the digits 0 and 1 that computers use to process data. The computer user speaks into a microphone, which translates the words into a continuous electric current. Other devices break this current up into individual pulses of current. The computer uses the software to translate the pulses into 0’s and 1’s.
Some software can translate a small number of simple words spoken with a wide variety of speech patterns and accents. This type of software is common in telephone-directory assistance systems. The development of systems that can handle long words and a large vocabulary is one of the largest challenges for the software industry.
